ildpil/text-anonymization-benchmark
收藏Hugging Face2025-04-11 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/ildpil/text-anonymization-benchmark
下载链接
链接失效反馈官方服务：
资源简介：
TAB数据集是一个用于文本匿名化的语料库，包含来自欧洲人权法院的1268个英语案件。这些文件经过人工标注，包括个人标识符的信息（包括它们的语义类别和是否需要遮掩）、机密属性和共指关系。
The Text Anonymization Benchmark (TAB) dataset is a corpus for text anonymization containing 1,268 English-language court cases from the European Court for Human Rights (ECHR). These documents have been manually annotated with information about personal identifiers (including their semantic category and need for masking), confidential attributes, and co-reference relations.
提供机构：
ildpil
